Creating a Social Media Research Agent for Trend Analysis

Muhammad B. · 2025-08-18T07:16:25.573Z

Hi team! I want to create a social media research agent... what it does is that it gets a request... i.e "search about the latest trends in AI and make me a content plan based on what's working and going viral platform: all" the platform it scrapes are X, Tiktok, Insta and Reddit I have attached the workflow that i have made so far... The error i keep running into is that it just scrapes from one platform and runs the entire workflow based on that... is there a way i can potentially get it to run in parallel? would really appreciate some help here

Creating a Social Media Research Agent for Trend Analysis | Lindy Community

15 comments

· Sorted by Oldest

Alon J.
·
·
Hey Muhammad, I got your email at Support@lindy.ai as well. Will take a look shortly 🙂
✅1
Mia B.
·
·
Following 🙂
Muhammad B.
·
·
Alon J. Just following up on this again 🙂
Alon J.
·
·
Hey Muhammad, apologies about the delay! It looks like the way you've set up what is scraped is based on a condition statement. If you're looking to scrape all of these, you could have another option in the platform router for 'all' and then it will just be one sequential path of Tiktok -> Reddit -> Instagram -> X. Let me know if that aligns with what you're thinking!
Loading attachments
❤️1
Muhammad B.
·
·
Ohhh nicee yeah! I didn't think about it this way Thanks alot mate Alon J.
🙌1
Muhammad B.
·
·
Alon J. sorry to disturb again brother... but now i'm running into data size limits... Any chance you can help me with this?
Loading attachments
Muhammad B.
·
·
Because that is really the main purpose of this workflow to be able to process large amounts of posts data and provide actionable insights from them
Alon J.
·
·
Hey Muhammad! Are you able to send over the task URL for this instance? Thanks!
Muhammad B.
·
·
Sure here you go: https://chat.lindy.ai/muhammad-burhans-workspace-2/lindy/social-media-researcher-68[…]stRunId=689ee96c33a20721242289db&task=68a40ebc3fcc7e74649d71c6 Alon J.
Muhammad B.
·
·
Hi Alon J. anything on this?
Alon J.
·
·
Hey Muhammad, apologies! My team is looking into it. However, there was a similar issue recently that was just fixed. Are you able to try again and let me know if you get the same issue? Thanks!
Muhammad B.
·
·
Tried twice, wasted a ton of credits but now it just stops in between: First run: https://chat.lindy.ai/muhammad-burhans-workspace-2/lindy/social-media-researcher-68[…]stRunId=68a6d0e2cb4e438d2432c035&task=68a6d0e2cb4e438d2432c03c Second run: https://chat.lindy.ai/muhammad-burhans-workspace-2/lindy/social-media-researcher-68[…]stRunId=68a6d0e2cb4e438d2432c035&task=68a6d338e841f66ba0e8e4d5 Alon J.
Alon J.
·
·
Ah yeah I see -- it looks like the LLM is failing on the backend. I'm curious if Marvin A. has an idea of how to resolve this. Otherwise, will need to send over to eng team to debug
👀1
Muhammad B.
·
·
Yeah would really appreciate support on this asap! I'm running late on a deadline here... so would really appreciate it if it can be resolved soon Alon J. It is essential for me to continue my Lindy Plan and use it across the board for all my clients as well
Marvin A.
·
·
you’re hitting a token limit with all the llms. so it’s just too much context you’re trying to pass. you’re currently at 1.5 million tokens, which is way too much for even the biggest ones. so i’d either recommend using better filters or trying to batch these and like use the clear task context action to filter for only most relevant results. Hope this is helpful. Alon J. Muhammad B.
🙏1

Alon J.
·
·
Hey Muhammad, I got your email at Support@lindy.ai as well. Will take a look shortly 🙂
✅1
Mia B.
·
·
Following 🙂
Muhammad B.
·
·
Alon J. Just following up on this again 🙂
Alon J.
·
·
Hey Muhammad, apologies about the delay! It looks like the way you've set up what is scraped is based on a condition statement. If you're looking to scrape all of these, you could have another option in the platform router for 'all' and then it will just be one sequential path of Tiktok -> Reddit -> Instagram -> X. Let me know if that aligns with what you're thinking!
Loading attachments
❤️1
Muhammad B.
·
·
Ohhh nicee yeah! I didn't think about it this way Thanks alot mate Alon J.
🙌1
Muhammad B.
·
·
Alon J. sorry to disturb again brother... but now i'm running into data size limits... Any chance you can help me with this?
Loading attachments
Muhammad B.
·
·
Because that is really the main purpose of this workflow to be able to process large amounts of posts data and provide actionable insights from them
Alon J.
·
·
Hey Muhammad! Are you able to send over the task URL for this instance? Thanks!
Muhammad B.
·
·
Sure here you go: https://chat.lindy.ai/muhammad-burhans-workspace-2/lindy/social-media-researcher-68[…]stRunId=689ee96c33a20721242289db&task=68a40ebc3fcc7e74649d71c6 Alon J.
Muhammad B.
·
·
Hi Alon J. anything on this?
Alon J.
·
·
Hey Muhammad, apologies! My team is looking into it. However, there was a similar issue recently that was just fixed. Are you able to try again and let me know if you get the same issue? Thanks!
Muhammad B.
·
·
Tried twice, wasted a ton of credits but now it just stops in between: First run: https://chat.lindy.ai/muhammad-burhans-workspace-2/lindy/social-media-researcher-68[…]stRunId=68a6d0e2cb4e438d2432c035&task=68a6d0e2cb4e438d2432c03c Second run: https://chat.lindy.ai/muhammad-burhans-workspace-2/lindy/social-media-researcher-68[…]stRunId=68a6d0e2cb4e438d2432c035&task=68a6d338e841f66ba0e8e4d5 Alon J.
Alon J.
·
·
Ah yeah I see -- it looks like the LLM is failing on the backend. I'm curious if Marvin A. has an idea of how to resolve this. Otherwise, will need to send over to eng team to debug
👀1
Muhammad B.
·
·
Yeah would really appreciate support on this asap! I'm running late on a deadline here... so would really appreciate it if it can be resolved soon Alon J. It is essential for me to continue my Lindy Plan and use it across the board for all my clients as well
Marvin A.
·
·
you’re hitting a token limit with all the llms. so it’s just too much context you’re trying to pass. you’re currently at 1.5 million tokens, which is way too much for even the biggest ones. so i’d either recommend using better filters or trying to batch these and like use the clear task context action to filter for only most relevant results. Hope this is helpful. Alon J. Muhammad B.
🙏1